The Financial Econometrics of the Distribution of Mutual Fund Performance

Arnold Kling asks the following in Education vs. Money Management:

Consider the following two propositions, one in education (E) and one in finance (F).

(E) One way to improve education would be to get rid of the bottom 10 percent of teachers and to try to replicate more widely the techniques used by the top 10 percent of teachers.

(F) One way to improve financial markets would be to get rid of the bottom 10 percent of money managers and to try to replicate more widely the techniques used by the top 10 percent of money managers.

What is interesting is that I know many economists who believe (E), although they may be skeptical that the best teaching can be replicated. However, I do not know any economists who believe (F).

(h/t Karl Smith who has additional comments.)


Is (F) obviously false? As I read the question:  Will the bottom 10% of money managers do worse time over time?  It’s not obviously false. Remember that efficient markets – in the strong, Fama and Friedman sense of efficient financial markets – are not efficient because every manager pulls the same portfolio, they are efficient through arbitrage opportunities. If there are noise traders, smart arbitrageurs will cancel their positions out. They could be driven out of the market, though that isn’t necessary. If a manager is dumb, he or she could likely be dumb in the next time period as well, but that manager could be able to continue to attract investors for non-fundamental financial reasons.

This allows for a very strong efficient market theory that allows distributional consequences.   If there are, as the kids say, limits to how effortlessly arbitrage works in financial markets, or if noise traders introduce risks into markets that the assets don’t compensate, we can see certain “anomalies” in asset price movements.  Mutual fund or hedge fund managers who take advantage of said anomalies could get better performance if they lean into the anomalies, and worse if they lean against, though it isn’t clear what type of “skill” that is.

It’s been a while since I’ve been on the frontier of cutting edge research into the financial econometrics of the distribution of mutual fund performance. (I was so young back then.) I’m pretty sure this research survives into the Great Recession and financial market collapse.

Let’s go to a personal favorite, Mark M. Carhart, On Persistence in Mutual Fund Performance The Journal of Finance, March 1997 (my bold):

Using a sample free of survivor bias, I demonstrate that common factors in stock returns and investment expenses almost completely explain persistence in equity mutual fund’s mean and risk-adjusted returns. Hendricks, Patel and Zeckhauser’s (1993) “hot hands” result is mostly driven by the one-year momentum effect of Jegadeesh and Titman (1993) but individual funds do not earn higher returns from following the momentum strategy in stocks. The only significant persistence not explained is concentrated in strong underperformance by the worst-return mutual funds. The results do not support the existence of skilled or informed mutual fund portfolio managers….

This article does much to explain short-term persistence in equity mutual fund returns with common factors in stock returns and investment costs. Buying last year’s top-decile mutual funds and selling last year’s bottom-decile funds yields a return of 8 percent a year. Of this spread, differences in the market value and momentum of stocks held explain 4.6 percent, differences in expense ratios explain 0.7 percent, and differences in transaction costs explain 1 percent…

Although the top-decile mutual funds earn back their investment costs, most funds underperform by about the magnitude of their investment expenses. The bottom-decile funds, however, underperform by about twice their reported investment costs. Apparently, these results are not confined to mutual funds: Christopherson, Ferson, and Glassman (1995) reach qualitatively similar conclusions about pension fund performance. However, the severe underperformance by the bottom-decile mutual funds may not have practical significance, since they are always the smallest of the funds….

The evidence of this artcile suggests three important rules-of-thumb for wealth-maximizing mutual fund investors: (1) Avoid funds with persistently poor performance; (2) funds with high returns last year have higher-than-average expected returns next year, but not in years thereafter….

As of that paper, there is visible persistence in mutual fund performance.  Media loves to talk about killer mutual fund analysts who beat the market.  And this does exist.  Does that mean mutual fund managers have skill? What Carhart did was re-examine the results controlling for survivorship bias and, crucially, already known-anomalies within asset price movements. With that in mind he found that mutual fund managers add no value – all they do is replicate already known anomalies, that come with their own debate about market imperfections, risk-weighting, etc. as well as eat up a lot of costs, costs that cancel value added.

The only exception comes with the bottom decile and top decile. Crucially, the top decile doesn’t add value when you consider the extra fees they charge. The bottom decile does far worse consistently. Key chart:

On that evidence I would tentatively accept (F).  Stay away from the bottom barrel of money managers.

This entry was posted in Uncategorized. Bookmark the permalink.

6 Responses to The Financial Econometrics of the Distribution of Mutual Fund Performance

  1. Howard Wu says:

    Had you reframe the question leaving off the teacher part and said eliminate 90% of the managers, I think you would have gotten a more positive response.

  2. Ted K says:

    Seems a long time since you did a pure finance post. Terrific as usual Mike. In my opinion, although I disagree with you on politics and such, you are still the best blogger bar none on Finance and one of the best finance writers period. Up with guys like Seth Klarman.

    I would like to see you write more on the limited choices of 401k plans and how some businesses take kickbacks from the planners, and the only 15–20 choices of funds (and poor choices at that) in most plans are because of hidden kickbacks and basically, bribery. More people need to rollover their money to IRAs where they have more choices and can often make better choices than the “professionals” without the so-called “active management” fees .

    I forgot the name and timing of the rule, but seems around July or June this year many employees will be locked into and cannot withdraw their own money out of their companies’ 401k plan. I’m not sure how prevalent this is, but I know of at least one person with similar such circumstances this year. You need to get the word out on this stuff Mike.

    Also I think many elderly people are being conned and manipulated into reverse mortgages that are not to their advantage. If anyone can do justice to these two topics it is you Mike. PLEASE??

  3. Chris J says:

    I think the difference between the two is that money-management is a zero sum game, while teaching is not (unless you believe that the top 10% of money managers are actually improving the performance of the companies they invest in).

    When a child does better on a test, it doesn’t follow that some other child does worse on a test. But if a money manager buys a stock with an intrinsic value of $20 for $10, then some other money manager must necessarily be selling that same stock for $10 when it is worth $20.

  4. Walt French says:

    I won’t challenge your conclusion but think if you dabble in this particular black art, you should consider subtleties that could completely reverse them.

    1. Over the time period of the FFC chart, the “Small” and “Value” factors would’ve done quite well. (And really, over ANY time period, there will be some average return to ANY risk factor.) A manager who was clever enough to ex ante identify a style tilt that would out-perform deserves credit for that insight luck, but the attribution here strips it away.

    2. That top decile 0.6% —> 0.5%/month looks awfully tempting. I’m not aware of many mutual funds that charge in excess of 6%/year — certainly not any of the ones that I think are skillful. That said, those may have been pyrrhic alphas — out-performing lousy style positioning for a net below-index result.

    3. While money managers are measured on their investment results, they are selected and retained at least in part for the quality of the information they provide investors about asset allocation — which investing approaches should do well, etc. — and other services such as reporting. Reporting, marketing materials, the quarterly outlook and pricing add up to a non-trivial part of a small fund’s expenses. (Many investors get a real eye-popping education when they see their funds’ statements.) ;^>

  5. Walt French says:

    [pls add this sentence to #3] For investors with say, $50,000 in a fund, $250 worth of educational materials would be treated as 0.5% but ranking managers as if this information was worthless is wrong: I’ll wager that virtually all of the readers of this blog derive their income from selling information.

  6. Bkay says:

    Walt: I’m not pulling up the original paper (and don’t recall the precise details from when I read it), but I’m pretty certain that the average excess return given in the graph above is a return above T-bills rather than a risk-adjusted alpha over the 4-factor model. If it was the latter, there’s no way that the average excess return would be positive for all deciles of funds at all times. Nearly by definition, about half the alphas would be negative.

    I don’t think 6-7% above T-bills is all that juicy for the time period of the Carhart paper.

    Mike: Don’t know if you’re interested in looking at more of the mutual fund performance literature, but there is at least one recent paper you should be sure to check out:
    Barras, Scaillet, and Wermers “False Discoveries in Mutual Fund Performance”

    They use a pretty ingenious method of sorting out funds that would have persisting positive and negative return from those that have approximately zero alpha but get lucky over the time period in question. They even use the same method over individual decades to show how the percent of mutual fund managers with persisting positive alpha has shrunk over time. Finally, they show that the strongest persistance occurs in the bottom tier managers, who suck year after year.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s